Primary human xenografts are human cancer tissues which are xenotransplanted into immunodeficient mice. Numerous studies have demonstrated that these xenografts much more closely resemble primary human cancers at the histological and molecular level then do xenografts generated from cell lines propagated in vitro (f-3). Primary xenograft models have proven most valuable in studying tumor biology and cellular heterogeneity, as well as for evaluating the efficacy of therapeutic agents. A prime example of the utility of primary human xenografts has been In the identification and characterization of human cancer stem cells (CSCs). UMCCC investigators used these models to report the Initial identification of CSCs in cancers of the breast, pancreas, head and neck and ovaries (4-6). Subsequently, xenograft models have been utilized by other UMCCC investigators to investigate CSCs in the brain, liver, adrenal and lung cancers {9-11). In addition to identifying CSC populations In various tumor types, these models have proven invaluable in studying the pathways which regulate them and in developing CSC-based therapeutics. Indeed, agents targeting CSCs regulatory pathways including Notch, Hedgehog, HER2/Akt, Wnt and cytokine signaling are being evaluated In multiple tumor types by UMCCC investigators. These preclinical studies have already led to the development of several early stage clinical trials targeting these CSC pathways. An important component of the translational oncology research strategy involves the development and testing of molecular targeted therapeutics In these xenograft models.